Motion magnification algorithms for video-based breathing monitoring

In this paper, we present two video processing techniques for contact-less estimation of the Respiratory Rate (RR) framed subjects. Due to modest extent movements related respiration in both infants and adults, specific algorithms efficiently detect breathing are needed. For reason, motion-related variations signals exploited identify monitored patient simultaneously estimate RR over time. Our methods rely on motion magnification that enhance subtle respiration-related movements. particular, amplitude- phase-based considered extract reliable signals. The proposed systems perform spatial decomposition frames combined with proper temporal filtering information. After periodic (or quasi-periodic) respiratory extracted jointly analysed, apply Maximum Likelihood (ML) criterion fundamental frequency, corresponding RR. performance presented is first assessed by comparison reference data. Videos framing different subjects, i.e., newborns tested. Finally, accuracy measured terms normalized Root Mean Squared Error (RMSE), demonstrating superiority, performance-wise, method.